Some principles for reducing consumption in new media art and technology 🍔👨‍💻

In the past couple years I’ve been greatly influenced by the concept and community oriented around permacomputing.

There are huge environmental and societal issues in today’s computing, and permacomputing specifically wants to challenge them in the same way as permaculture has challenged industrial agriculture. With that said, permacomputing is an anti-capitalist political project. It is driven by several strands of anarchism, decoloniality, intersectional feminism, post-marxism, degrowth, ecologism.–Permacomputing wiki.

When I first heard folks online discussing this concept, my ears perked up. It was addressing an issue that has long simmered in my mind in art and tech spaces. It nagged at me when for exhibits people would buy giant flatscreens, then return them later after the show, perhaps resellable, but maybe not. Maybe they’d turn to trash. Or the constant purchase of other new computer hardware based on dubious trends (Amazon Alexa, internet-enabled product buttons, etc). The endless rush to consume more pixels, more speed…

I have an interest in reducing consumption so as not to increase plastics, heavy minerals and all the pollution associated with computing, and the production of new products. This means that rather than adopt the latest trends and expensive GPUs for ever-increasing levels of detail in 3d, VR, or new expensive proprietary headsets for closed-platform metaverse that will eventually get abandoned, I prefer to use recent or yesteryear’s tools, especially those that are more resilient. I’m thinking of tools that are better made, repairable, “open”, customizable, and work with reduced bandwidth, and offline, as a start.

I was particularly challenged when reading Permacomputing Aesthetics: Potential and Limits of Constraints in Computational Art, Design and Culture (PDF) (Mansoux, Barok, Howell, and Heikkilä) to begin to brainstorm what kinds of constraints in consumption and computation can be generative and help spur on creativity and a flourishing of environmentally-minded approaches to art and code and community.

A couple of personal experiences has made this seem possible, maybe even delightful and revelatory.

First, the absolutely thriving retro scene that has largely solved emulation of mostly all 80s/90s game consoles, operating systems, software is huge. So many tens of thousands of programs largely still work fine, sometimes are better than the latest thing, and work through emulation. Generally I find them personally fun and useful. My students, not alive in the original era of these consoles and systems also find them personally meaningful and useful. Here I’m thinking of running software and games on DosBox and MAME emulators, for example. And using Hypercard and firing up old “stacks” collected from the Internet Archive as just a few examples. There is so much software and games out there, one could be satisfied just with all of this.

Secondly, a lot of the old hardware from this era largely works, and works well, or can be revived. And the use of physical game carts, despite being retired or forgotten by the large companies that originally made them, work long after operating system updates, lost disks, lost files, and made-defunct computers, phones, and apps have ceased to make their software systems available. You can find tens of thousands of old game cartridges, old computers and the like on ebay or elsewhere, that still work. And the collector community lovingly preserves and creates new hardware to protect, copy and save and make useful old software and hardware. New add-ons, or replacement parts are common, or add new functionality.

Emulation and preserved software and hardware isn’t the only needed path forward. I’ve been heartened in the past decade by the tool and ecosystem based on Pico-8, originally released in 2015, a game engine that is so seemingly perfectly designed and refined, pointing to a new path forward. The community around it is so strong, as to provide almost enough on its own to make a complete game system, as long as you appreciate its aesthetic. It’s a successful model of the ecosystem I’d like to see more widely. With a small application, limited screen size, limited palette, sound engine, sprite graphics editor, and code editor all-in-one, you can write endless games, with “productive constraints.” There is an endless stream of shared, open games from people around the world. All games can be saved as carts, uploaded or distributed through the application’s network-enabled browser. And there are thousands upon thousands of fun games to be played, with new ones constantly coming out. There’s also a bulletin board where people discuss, share, and collaborate. And there are a number of open source fantasy consoles inspired by Pico-8 with different productive constraints, tools and games to play. These all work fab on computers with limited specs. No special GPU or high-end components needed. The creator of Pico-8 has been working for the past two years on a complementary “fantasy workstation” called Picotron, that allows writing system software, games, and applications on the friendly Lua language, the same as Pico-8 and many other fantasy consoles. As with Pico-8, its more wide but productive constraints and design considerations allow it to work on a range of computers even on the low end. And software written on any computer running Pico-8 or Picotron can be ensured to run on any other, regardless of operating system and system specs.

I keep emphasizing low end and older computers because not being forced to buy a new machine means we can consume less, produce less. When my Dell XPS laptop had a failing battery a couple years ago, I had it replaced. When the audio port cracked and separated from the laptop from a fall this spring, I had the plate replaced, and my computer has felt as good as brand new since. Running an open source Linux operating system as opposed to the bloat of Windows or the locked-down Apple Mac means my Linux laptop is as zippy and light as ever, even at over 5 years of age, and I hope to be able to use it for many years more.

While the web has in many ways gotten worse (see my last article for some complaints: pop-ups, ads, machine-learning-written garbage, clickbait, cookies, trackers), it’s also become a “platform” of epic proportion that can serve well on low-end machines if keeping within certain bounds. While not ignoring some of the previously mentioned downsides and others such as bandwidth-heaviness, security issues, it is also a place where millions of web applications can be found and distributed. You can also “Save As” and download some website applications depending on how it was made, which should become a new area of focus. And applications and reading and writing-oriented experiences most obviously work extremely well. One could in fact design with HTML/CSS (maybe even some Javascript) to design minimal, useful offline personal applications outside the standard act of making online, networked apps and websites. This is for example Electron apps in a nutshell, but other approaches that don’t require downloading a browser for each application could be taken as well. For examples of applications and approaches in this style, I especially like the work of Omar Rizwan. Check out his TabFS for an example of software that allows one to use the browser as a personal tool-making system to modify pages, applications and other things online and offline. There are a number of websites, books and even StackOverflow answers with suggestions for building offline-first browser-based applications.

In order to serve the need to make long-lasting, less wasteful, reduced consumption, I think we also have to reject the gate-keepers, the “priesthood” of computer-programmers, as described by pioneering computing engineer Lee Felsenstein. We need to advocate for learning, wielding, and building our own tools, collaboratively with others where possible, and using tools one finds comfortable. One of the important tools and languages I learned to program with is Scratch, oriented around child learners, an inheritor of the methods and goals of LOGO. I made my own games and used Scratch as I started to learn to program with Processing. There were things I couldn’t yet make with the text-only languages that Scratch allowed me to do. When I was in middle school we learned to make applications, and interactive stories and games in Hypercard and Hyperstudio. These tools with visual interfaces (see also Klik n Play, Decker, Interface Builder, Quartz Composer, Twine, Max, among others) are a key way for people to build useful “stuff” and tools. Spreadsheets are another way. With this approach, personal tools can work far into the future, and be accessible for more uses than the limited nature and surveilled “apps” in a commercial, proprietary “app store” for example.

Another way of saying this is that we need to embrace a kind of “amateurism” in DIY spirit. Expertise doesn’t need to be eliminated, but it can withstand the work of the amateur community. Every person a guide and a teacher.

Open source and similar approaches to building user-serviceable, repairable, software and hardware is key. If we have closed car computers, washing machines, thermostats, and lights, then when the company moves on to the next thing, forces someone to upgrade to purchase more, changes their pricing, and goes out of business when the startup fails, who is stuck holding the (useless) bag?

Hand in hand with reducing reliance on closed-source, non-repairable tools, and online streaming services and platforms is using just the right amount of computer technology by using less. This will depend person to person, but we are infatuated with the latest tech, and that tech is contributing to climate change. Stress offline over internet-connected, streamed and controlled. Instead of a car to get to the show, use a bike. Instead of an app to get into a building, use a key. Instead of purchasing more monitors and projectors, can we reuse what we have? How many times will you need to buy the newest hardware, update your operating system, or purchase a new phone? Recently at the Whitney museum I saw an artist I recognize with a wall label explaining one had to download a particular app to see their AR artwork in the museum. How many visitors walked right past that? Over 99%. In 3 years, will the app still work? Likely not! These things are convenient until they’re not, and at too high a cost.

Lastly, it’s important to say that while this has so far sounded prescriptive, we should approach things in an exploratory, open way. When we need to use certain new hardware, purchase a new printer or projector, for example, we can do that. And we are still learning and devising new approaches and considerations in climate change, consumption and computing. It’s just that the default today as I write this is new and AI-enabled and proprietary, non-repairable, “app”-oriented. Especially with LLM models from companies like OpenAI, we are consuming ever-increasing power, expanding non-renewable fuel sources and factories. Companies are demanding building new nuclear power plants. This is madness, and a sign of how extreme consumption has become normalized, at the expense of our own and planet’s health.

Rather than enjoying being a scold, I hope to impart there are opportunities here and genuine enjoyment at taking another path. I’ve pointed to a few, and there are likely many more approaches we can develop in community. As an artist working in new media with code and software, there’s a lot I can personally do to be mindful of over-consumption, and it starts with being mindful of of my own consumption, talking about it with others, and modeling behaviors. I don’t think I’ve touched on all considerations or found all the opportunities for reducing consumption, and I think the challenge of Permacomputing Aesthetics (PDF) paper is one of opportunity to map out new approaches, movements, and aesthetics to come.